In C++ this is done for you when control exits the enclosing scope of a variable.
Only for local, stack allocated objects. You still have to call the destructor's manually for heap allocated or global objects. Which doesn't mean you can't do RAII in C++, it just means you have to understand semantics of the language you are using. But then, this is true for any style of programming, in any language. Admittedly, it is much easier to implement most styles in a language such as lisp or smalltalk (which provide rich introspection and little heavy handed bias) than in a language such as java or C (with weak introspection and strong built-in style assumptions)--but "less easy" is not equivalent to "impossible".
You can use RAII in any real language. You can write structured, functional, or object oriented programs in any real language as well. Conversely, you can write unstructured, spaghetti muddled dreck in any language. The semantics of the language don't matter nearly as much as how well you use them and how rigorously you apply your chosen style.
There's nothing to stop you from writing your own, and calling it yourself just as you would in a manual-allocation system.
Think about it--someone could write a C library to use an underlying garbage collection system and on semantic grounds you'd have no way of knowing it, would you? You'd release your memory, the library would mark it as "dead" but otherwise do nothing, and the GC would get around to recycling it when it felt like it. For that matter, the GC might even be in the OS, hidden from the language itself. How would you ever know?
With that in mind, why does the fact that you know about it change things?
--MarkusQ
P.S. You aren't guaranteed that a destructor will be called in a manually managed system, either. If it was guaranteed, the system wouldn't really be manually managed, would it? And nothing in a GC language prevents you from doing exactly the same things you otherwise would have been forced to do.
aside from the performance issues, predictable memory management can be used for controlling access to files and similar resources, creating safer thread locking code and even providing better error messages.
This is silly. None of these have any connection to garbage collection; you can write "destructor's" in a garbage collected language, and do everything in them just as you would have in a non-GC language.
The advantage comes from the RAII style of coding, not from the absence of a garbage collector. In fact, most modern GC languages provide better RAII support, in that there is no way to get an uninitialized object object.
In a way, it reminds my of the old bounds-checking arguments. A fair number of people used to resent/resist built in bounds checking, with very similar arguments (performance, trusting the coder, illogical correlations between manual bounds checking and various Good Things(TM), etc.) and thus we still to this day struggle with buffer overruns and related problems.
I think what the parent poster meant by "you're done" was, the entire area is made unlivable for hundreds of years. There isn't any second chance.
Yes, I realize that is what he meant. My point is that it simply isn't true. 30 years is a more realistic estimate, but even that is unrealistic. The actual risk is far less than 1% increase rate of premature death (mostly due to an additional 0.005% chance of thyroid cancer) which is itself far lower than many risks people take daily. Living with a smoker, for example, is more dangerous than living just outside the Chernobyl wildlife refuge.
This is much worse than a single person or family accidentally killing themselves.
Much worse for whom? Surely not for the single family. To the dead it really doesn't matter now, does it? And for the living, although there is a greater emotional impact to a hundred people dying of cause A in one event than there is for thousands dying of cause B in a thousand separate events, it isn't a very rational position to take.
The same goes for dangerous places. One big dangerous place is far better than a thousand small ones, if only because it is easier to avoid.
Then you see the problem: a "normal fire, not unlike accidents we commonly see in the fossil fuel industry" was enough to render a huge area of land permanently uninhabitable and ruin thousands of peoples' lives.
Except that the area wasn't made "permanently uninhabitable" (but even if it was, that isn't such a big deal when you consider that 80% of the Earth's surface is perminantly uninhabitable to the same or greater degree). At worst, it's not quite as safe as it was, and won't be for another 30 years or so.
And as for ruining people's lives, it's a drop in the bucket compared to the use of fossil fuels (or the use of tobacco, or wars over oil, or the manufacture of pesticides, or...). To cite one such number out of context is simply FUD. Nuclear, per kilowatt hour, is still safer than the alternatives, is more scalable, and has less environmental impact.
That is why so many people are uncomfortable with nuclear power, and seek a better alternative.
I agree that the beliefs you outlined are responsible for the widespread rejection of nuclear power. But that doesn't make the beliefs sound--reality isn't a democracy, and many people believing something doesn't make it true.
(btw, I think the public's mind blames hydrogen more than helium;^))
Oops. Typo. Thanks.
I think that "nuclear or coal" is a false choice.
Agreed. I'm a fan of space-based solar myself, but that's a topic for another day. The main advantage of nuclear is that it is available now, and is the only technology we have that is far enough along to prevent world war three from erupting over control of the oil. Conservation, wind power, etc. just don't have the scope to prevent that deadly showdown.
It doesn't matter that there were no major health issues caused by Three Mile Island. It is still the primary reason for the souring of opinion toward nuclear power in the US because it was a mistake that nearly ended in disaster.
So the facts don't matter, the scare story is what counts. You do realize, of course, that almost anything that happens could be characterized as having "nearly ended in disaster" if you're willing to play fast and loose with the facts. So all you're really saying here is that the scare stories worked, which was my point.
It still exhibits the fundamental problem of nuclear power: one mistake and you're done. People do not believe it is wise to try to operate a facility with those kinds of tolerances.
Do these people by any chance drive cars? Because, you know, one mistake while you're driving and you're done. Same thing with using a fork.
Something can be very safe, and still be considered too dangerous to use.
Do you mean that people can (incorrectly) decide--or be persuaded--that something is too dangerous even though it is in fact safe? Or are you mired in some sort of Orwellian double speak, where "Safety is danger"?
For example, we don't use hydrogen in lighter-than-air vehicles anymore. It can be done safely for significantly less money than helium, and there were very few accidents. But you make a little mistake and "Oh the humanity!".
Excellent analogy to Chernobyl, which was after all a normal fire, not a reactor problem, and not unlike some of the accidents that we commonly see in the fossil fuel industry. The Hindenburg disaster didn't have anything to do with the hydrogen, though it was widely blamed; the fault lay with the coating applied to the fabric and the same thing would have happened if it had been filled with helium. But even though this fact has been known for years, in the public mind it will always be helium's fault.
It doesn't really matter that you only have a 1 in 6 chance of death in Russian Roulette. Eventually, you get a bad pull.
Not at all true. If you are engaged in other, risker activities your chances of death from the Russian Roulette go down. If, as in the case of nuclear, the risks are far lower than (for example) driving a car, your chance of death isn't effected much at all.
And the truth is that no other form of energy can have such a long-lasting or damaging disaster.
Tell me that when your front yard is hot enough to cook a pizza--or right, we'll all be dead then, won't we. I think the scare stories of both global warming and nuclear disasters are overblown, but to say that nuclear is worse is absolutely nuts. Read about surface conditions on Venus (which is very much like Earth, except it had run-away global warming and we haven't yet).
The other thing to consider: A nuclear disaster is something that would represent the most extreme failure mode of the system, while global warming is a natural consequence of the normal operation of the fossil fuel economy.
Hydroelectric dams come close, but you don't have to deal with millions of acres of radioactive land, just death and mud.
We have millions of acres of radioactive land right now. Pretty much anywhere there's granite is radioactive, for example. It may come as a surprise to you, but we get the radioactive material we use for nuclear power by digging it out of the ground. It's been there all along, and if we don't dig it up it will simply stay there, making the ground radioactive, just as its done for billions of years.
What I find ironic is how often people who don't trust the fossil fuel industry, and claim not to believe anything they say, etc. have been taken in by the anti-nuclear FUD spread by the very people they claim to distrust.
It's like some bad comedy routine.
Joe Public: I don't trust you. Coal and Oil guy: I can understand that. Joe Public: Nothing you can say will make me trust you. Coal and Oil guy: I know just how you feel. Joe Public: You do? Coal and Oil guy: Sure. See that guy standing over there? The one with the pocket protector? Joe Public: What, Nuclear Guy? Sure, I see him. Coal and Oil guy: I don't trust him at all. Joe Public: Why not? Coal and Oil guy: He wants to kill all our babies and make giant insects and stuff. Joe Public: Really? Coal and Oil guy: Really. And he wants to make stuff that will kill people a bazillion years from now if they so much as think about it. That's why I don't trust him. Joe Public: Wow. Thanks for the warning. But this isn't going to make me trust you any more than I did before. Coal and Oil guy: I can understand that. Just so long as you don't trust him either. Joe Public: Or don't worry about that. That guy is scary!
I've changed the topic heading b/c this is not a discussion about right or wrng - its damn interesting.:-)
...the "matrix" (that fucking movie ruined the term!!!)...
Back to the discussion:
There are a few problem with the "bouncing back and forth in time" explanation:
We should see equal quantities of matter and anti-matter
The amount of "memory" the magic particle would need would be astronomical (not a killer, since there's only one of them, but it seems inelegant)
It conflates thermodynamic time, casual time, and time-as-a-dimension in a way that seems less than helpful here
The last point is the real kicker. If you allow the backward flow of information along world lines it seems to me that you loose the ability to compute intervals, and your whole space-time metric falls apart.
(I'm aware that there are some "interactionist" models that accept this, but I don't recall much about them other than the assumption that the information flow across a unit of area is strictly limited to preserve the gross properties of the macroscopic world (e.g. we can't drive to the last century for the weekend) which would again give you the ability to measure something without affecting it. And thus even in them, there is unidirectional causuality of a sort, even if it isn't tied to a well-defined space time.)
If you don't allow backward flow of information, then being "strictly after" an event (delta_t > 0 and delta_t^2 > delta_x^2) should make it possible to measure it without affecting it.
If Schroedinger's Cat were conducted, the cat would not "be both alive and dead" in any meaningful sense.
In your words "The whole copenhagen position sounds good at first, but after a few decades of consideration, it stinks."
Now back to our regularly scheduled debate
you would in effect be claiming that all measurement was impossible
an accurate measurement of anything IS impossible. WE already know that.
You are misquoting me right after quoting me verbatim. My point was (and is) not that you would in effect be claiming that infinitely accurate measurement was impossible, but that all measurement was impossible. The position is logically untenable precisely because it reduces to an absurdity.
because the speed of light will prevent any influence from the thermometer on the sample.
actually, you're making an assumption that strict causality exists. It has never been proven, and is an unworkable assumption because it then requires artificial constructs such as superpositions of states.
Say what? Please explain how, even in the absence of strict causality, when I measure (via spectral analysis) the temperature of a star that no longer exists (because, unbeknownst to me, it went nova millions of years ago) the temperature of my instruments can affect the temperature of the star, That doesn't sound like QM to me, it sounds like voodoo.
BTW, you misquote me when you say:
You said that of Schroedinger's cat (which asserts that macroscopic objects enter into a superposition of states) "Sounds weird, but its already been proven to be the case"; to now claim that this is "a bit of a brainfuck" while still maintaining that it is a proven, real effect doesn't wash.
The 2-slit experiments I'm referring to neither require nor use a mirror. You're referring to something else.
The two-slit experiments you are referring to do not put (or claim to put) macroscopic objects in a superposition of states (which is what Schroedinger's cat is all about). So they don't prove (as you claimed) that this actually happened. The only experiments to date that I am aware of to actually test the issue did use mirrors (actually, a metal grid that served as a mirror at the wavelengths in question), for the reasons I outlined.
As for the temperature, there is no way that a thermometer of any size can't affect the sample its measuring unless it already is at the same temperature as the sample.
This really has nothing to do with QM (except in the popularized explanations), but in any case it both misses the point and is wrong.
For a clear example of how it is wrong, consider a thermometer that works by passively measuring the emission spectrum of the sample; in that case, it needs to be at the temperature of the background it is obscuring not at the temperature of the sample. But if it is far enough away even that won't matter, because the speed of light will prevent any influence from the thermometer on the sample.
To see why it misses the point, consider that, if you really meant it, you would in effect be claiming that all measurement was impossible. The real question is, does the thermometer affect the temperature enough to in turn affect the measurement it is making? This is a real question, of practical significance in many situations, and is the only part of all this that touches on QM. In the classical realm, you can always contrive a finer, more devious way to make the measurement to the desired accuracy while keeping the disturbance below the threshold. The key point of QM is that you can't do this indefinitely, that there is a quantifiable, universal limit to what is knowable without regard to any such "disturbance" by the measurement and thus not thwartable by clever design of instruments.
Every measurement, even in the macro universe, affects the thing measured. That this would NOT be the case at the quantum level, while counterintuitive at first, would upon reflection be surprising.
Not strictly true (for another example, I can attempt by various means to measure the distance Alexander the Great traveled during his life, but I cannot in any way affect it), but in any case this isn't QM, it's just an analogy used (misleadingly, IMHO) to explain QM (specifically the Copenhagen interpretation) without getting into the math.
Also, there is no need to resort to a superposition of states (which has always been a bit of a brainfuck imho) when there are better models. Superposition is only required if the time scale is unidirectional and can't be "rotated out" of the question and replaced by another vector, which has never been shown to be the case. The copenhagen gang lacked sufficient imagination to see the obvious.
There may be no need to resort to superposition in general, but the specific point in question here, "Do macroscopic objects, such as cats, enter into a superposition of states as claimed?" can not be dodged in this way. You said that of Schroedinger's cat (which asserts that macroscopic objects enter into a superposition of states) "Sounds weird, but its already been proven to be the case"; to now claim that this is "a bit of a brainfuck" while still maintaining that it is a proven, real effect doesn't wash.
-- MarkusQ
I think "they" got hold of the same list.
on
DMCA Abuse Widespread
·
· Score: 2, Funny
There are four boxes to defend liberty with: the soap box, the jury box, the voter box and the ammo box. Use in that order.
I think "they" got hold of the same list, and their response plan went something like this:
Soap box -- let's consolidate all the news outlets in the country under a few corporate trees. Then fire a few warning shots at any reporters who don't get the memo.
Jury box -- No prob. Push the right of appeals so that everything can go to the SCOTUS. No jury there, and we can pack it with corporate-friendly judges by distracting the sheeple with some made-up hot-button issue. What do you think, evolution? Or should we stick with abortion?
Ballot box -- Check with Steve on how that e-voting thing is coming along.
Ammo box -- Before we write this one off, is there any way we can get our margins on domestic arms sales up closer to what we make selling the big stuff over seas? I know we can shut 'em down with the Patriot stuff, but I just feel like we'd be walking away & leaving money on the table.
I always had a problem with that experiment - it implies that human intellegence has some link with the state of the universe.
Sounds weird, but its already been proven to be the case
Actually, it has been found not to be the case. The experiment essentially uses a mirror instead of a cat, and either breaks or doesn't break the mirror. A mirror is used because (using an other aspect of QM) you can distinguish a mirror that is in a superposition of states (broken | unbroken) from one that is definitely in one or the other without gaining any information about which state it's actually in.
It turns out, the cat is always either alive or dead, the mirror either broken or unbroken. Macro scale objects do not enter into a superposition of states in the real world. The assumed reason for this is that the larger the object, the more fragile (speaking very loosely here) a superposition of states is. By the time you get to a macro scale object at normal temperatures, you will never get a superposition--just moving an electron somewhere in another galaxy would be enough to collapse it.
On the other hand, the idea of macroscale objects in superposition is not nearly so fragile. It seems to be entrenched (paradoxically) because it's so counter intuitive.
--MarkusQ
P.S. As for the rest of your explanation, it's right in spirit but wrong in fact. Operating at the same level of analysis, you could just keep making smaller and smaller thermometers, or playing clever tricks with them to measure the temperature of the water indirectly, or without affecting it as much, so the argument as you present it begs the question.
The whole point of QM is that there is a bottom to all this, a point where you can't build a smaller thermometer, and no combination of clever tricks will help you. Your argument only seems to explain this because you silently assume it from the very beginning.
They aren't worried that RF noise will set off the explosives; they are worried that RF noise will interfere with the equipment they use to set off the explosives. They don't use a long roll of wire and a plunger style generator anymore--they set 'em off by radio. And they'd rather you didn't set them off by radio before they were ready.
Competition is good for consumers (and the market) under the assumption that they can make rational, informed decisions about the competing products. The whole purpose of the cell phone style pricing schemes (as with advertising, complex tax codes, etc.) is to make rational comparison so difficult as to be practically impossible, and thus invalidate the assumption; they are an attempt to subvert the process.
There were different rates for peaks hours and off-peak hours, so that when you were feeling cheap you take that into account
You had a variety of rate plans to choose from (say, some with a base of pre-paid hours, or others with "anytime" hours) and so forth so that you could choose a plan to suit your needs
Some games came bundled with a computer, or you had the option of getting a new computer "free" if you signed a long term contract with a game
Every game came in two dozen flavours, all basically the same but with a non-systematic assortment of the features enabled or disabled
There were different payment plans available for new players, existing players, active players in other games, etc.
Some payment plans cost more than you wanted to play, but came with a piece of paper that, if you filled it in exactly right and sent it to the right address, on the right day, you might get most of your money back.
The payment plans were only available for a limited time, and the next time you looked into it they would have subtly changed.
If I'm right, you may want to go cell phone shopping.
For my self, I far prefer a company that simply tells me what their product costs and then lets me decide if I want it.
The idea of a test being repeatable does not require "moving forward in time, and seeing the results"; it only requires that independent observers can replicate the results. Take astronomy, for example. If I have a theory that says "if you look carefully at a region of space that meets this specific set of criteria you should find more of this type of object than of this, by a ratio given in this formula" I can test it by going out and performing the observations. You can confirm it by doing the observations yourself, on the same volume of space or on another of you choosing. Neither of us has to have a Universe-Rewind button; we don't have to reboot the galaxy, or wait for a new one to form.
The same logic applies to geology, cosmology, etc. as well as biology, of which evolution is a keystone.
--MarkusQ
P.S. In any case, the test you propose ("A proper test of evolution would be to have some species, propose they will evolve a certain way, and then verify that they do indeed evolve that way.") has been done many times and consistently turns out to confirm the theory. So even if I agreed with your artificially constrained definition of "test" the original point still holds.
Think about what happens when they take a step...
on
Cow Tipping is a Myth
·
· Score: 2, Informative
If they start to step by raising a leg on the far side, it will make them easier to tip.
If they raise a leg on the near side and try to step away, it will make them easier to tip.
Their best (and typical) response is to raise a near leg and move it towards the tipper, broadening their base and lowering their center of gravity. That, and only that, would makle them harder to tip. But that does not mean I agree with the articles conclusions.
Evolution is testable, falsifiable, and even demonstrable.
Evolution makes some rather stringent predictions about the sorts of plants and animals we should expect to find, and the ones that we shouldn't. For just one example, we should expect to see animals that are systematically willing to die for a chance to mate, but not for something to eat or drink. And that is in fact what we do see. There are hundreds of such predictions, and they have all turned out to be correct,
Darwin himself pointed out several ways in which his theory could be disproved, and many others have been discovered since. Yet for all the ways it could be falsified, it has not been.
We can actually see evolution happening, in everything from domesticated animals to germs. Why do you think doctors worry about overuse of antibiotics?
"First off, yes, it is a crime to leak the name of an undercover agent (or any other classified information) regardless of intent."
No, it is not.
Yes, it is. It's called "mishandling classified information" and you don't have to even leak it. Hiding notes about classified information in your socks suffices (remember Berger?). The whole point of having a system of classified information is that the people entrusted with the information aren't supposed to tell unauthorized people about it. Whether their intent is to impress their girlfriend, or burn a political adversary, or win a prize on a radio quiz show, or just make smalltalk while waiting in line at the theater doesn't matter.
As for the rest of your post, you might want to sniff the kool-aid you're drinking before you start calling names.
Now this administration may be able screw up the invasion of the wrong country...
As opposed to Bill Clinton's invasion of two "wrong" countries Haiti and Kosovo... a "quagmire" I think we're still stuck in. Not to mention his poor execution of the efforts in Somalia and his indifference to the people of Rwanda.
I am no fan of Clinton, but you can't seriously be comparing the scope of Haiti and Kosovo to Iraq.
leak the names of CIA agents...
At this point unfounded speculation at best. Besides, it's not a crime to "leak" the names of CIA agents unless the intent was to expose them. In this case, it was hardly the intent to expose an undercover CIA operative (which Mr. Wilson's wife was not), but simply to disclose how Mr. Wilson got the assignment. But speaking of breaking laws, who was it that lied to a Grand Jury abou a blow job? Oh yes, that was Bill Clinton.
First off, yes, it is a crime to leak the name of an undercover agent (or any other classified information) regardless of intent. And yes, despite the administrations carefully worded talking points she was undercover, and the information was classified as "secret" in the memo the CIA provided to the WH.
And the "intent to disclose how Mr. Wilson got the assignment" fib has been sunk by the time lines--unless you are claiming they began an organized campaign to clarify a statement three weeks (mid June) before the statement was made (early July) and are intending to split hairs about the distinction between why Wilson in particular was send (selected by the CIA, after being suggested by his wife) from the real question of why anyone was sent on this particular assignment (do obtain more information, as requested by Cheney).
As for the "unfounded speculation" aspect, you may want to catch up on the news. We now know that the administration has repeatedly lied about this issue, including the claim that Rove & Libby had "nothing to do with it" which was changed to "first heard about her from reporters" and then to "were acting alone, not as part of any organized campaign" and that they were doing it "in response to Willson's NYT opinion piece" but started weeks before the piece was even written and did so in an amazingly unified and coherent fashion. We were told that "Cheney knew nothing about it," even though today we learn that Libby's hand written, dated notes of a meeting with Cheney in the days before the campaign started include the salient details.
But I guess all this overshadows the fact that the 9/11 commission says Mr. Wilson lied about the Nigeria-Iraq connection, which is what the liberals want.
Stripping the political baggage from your statement (facts don't care who "wants" them), Wilson was disputed on a single point; he said he "saw" that the documents were forgeries, but had not in fact personally "seen" the original documents. Understandable as a miscommunication, and hardly discrediting, especially as (IIRC) he clarified the point as soon as he was called on it. He has been proven correct and Dick "We know they have WMD" Cheney has been proven incorrect on every substantive point.
--MarkusQ
P.S. For the record, I was up in arms about Clinton and the BJ too. Both for the perjury and (perhaps more importantly) for the effect on his family. Hillary can take care of herself, but imagine the effect that must have had on his daughter. Not to mention that the Democrats would have been up in arms, crying sexual harassment if a CEO or the president of a university had done something similar.
But just as I hold the Democrats responsible for their actions, I expect the leaders of my own party to behave themselves in a way that brings credit, not shame, on the party. And this cabal of nincompoops is doing more damage to the Republicans than any Democrat could dream of doing.
Just as I turned in disgust from the latest Dan Lyons OSS flamebait (nominally about MySQL but mainly about his fear and loathing of Open Source) I thought to myself "What to I want to read now to get the taste out of my metaphorical mouth?"
I Dvorak flamebait? No, we just had one of those, and anyway reading two flamebaits in a row is bad for my blood pressure.
I know what would hit the spot! A Roland Piquepaille troll! We haven't had one of those for a while.
It's as if the gods heard my plea.
Darn those gods to heck anyway. I never thought I'd say it, but I miss Jon Katz!
I think we disagree about who is the victim. It isn't the person who's computer is taken over (I would call them an accomplice by virtue of negligence), it's the people against who the bot net was used.
Look at it this way; if there was a sudden fad for leaving loaded guns on the roof of your car when you parked, and street gangs were using them to commit crimes, would you just consider the people who left the guns on their cars victims? Their guns were stolen, after all. Or would you say that they shared some of the responsibility for the crimes?
Slashdot Mirror
Hrm, did you notice that Firefox 1.5 is crashing as well on this exploit?
No, I haven't noticed that. Can you back up the claim? Or are you just blowing smoke?
--MarkusQ
In C++ this is done for you when control exits the enclosing scope of a variable.
Only for local, stack allocated objects. You still have to call the destructor's manually for heap allocated or global objects. Which doesn't mean you can't do RAII in C++, it just means you have to understand semantics of the language you are using. But then, this is true for any style of programming, in any language. Admittedly, it is much easier to implement most styles in a language such as lisp or smalltalk (which provide rich introspection and little heavy handed bias) than in a language such as java or C (with weak introspection and strong built-in style assumptions)--but "less easy" is not equivalent to "impossible".
You can use RAII in any real language. You can write structured, functional, or object oriented programs in any real language as well. Conversely, you can write unstructured, spaghetti muddled dreck in any language. The semantics of the language don't matter nearly as much as how well you use them and how rigorously you apply your chosen style.
--MarkusQ
There's nothing to stop you from writing your own, and calling it yourself just as you would in a manual-allocation system.
Think about it--someone could write a C library to use an underlying garbage collection system and on semantic grounds you'd have no way of knowing it, would you? You'd release your memory, the library would mark it as "dead" but otherwise do nothing, and the GC would get around to recycling it when it felt like it. For that matter, the GC might even be in the OS, hidden from the language itself. How would you ever know?
With that in mind, why does the fact that you know about it change things?
--MarkusQ
P.S. You aren't guaranteed that a destructor will be called in a manually managed system, either. If it was guaranteed, the system wouldn't really be manually managed, would it? And nothing in a GC language prevents you from doing exactly the same things you otherwise would have been forced to do.
You can't do RAII in GC languages because you've no control over when the destructor is called.
That's nuts. Just call it yourself, exactly as you would in a non-GC language.
--MarkusQ
aside from the performance issues, predictable memory management can be used for controlling access to files and similar resources, creating safer thread locking code and even providing better error messages.
This is silly. None of these have any connection to garbage collection; you can write "destructor's" in a garbage collected language, and do everything in them just as you would have in a non-GC language.
The advantage comes from the RAII style of coding, not from the absence of a garbage collector. In fact, most modern GC languages provide better RAII support, in that there is no way to get an uninitialized object object.
In a way, it reminds my of the old bounds-checking arguments. A fair number of people used to resent/resist built in bounds checking, with very similar arguments (performance, trusting the coder, illogical correlations between manual bounds checking and various Good Things(TM), etc.) and thus we still to this day struggle with buffer overruns and related problems.
--MarkusQ
I think what the parent poster meant by "you're done" was, the entire area is made unlivable for hundreds of years. There isn't any second chance.
Yes, I realize that is what he meant. My point is that it simply isn't true. 30 years is a more realistic estimate, but even that is unrealistic. The actual risk is far less than 1% increase rate of premature death (mostly due to an additional 0.005% chance of thyroid cancer) which is itself far lower than many risks people take daily. Living with a smoker, for example, is more dangerous than living just outside the Chernobyl wildlife refuge.
This is much worse than a single person or family accidentally killing themselves.
Much worse for whom? Surely not for the single family. To the dead it really doesn't matter now, does it? And for the living, although there is a greater emotional impact to a hundred people dying of cause A in one event than there is for thousands dying of cause B in a thousand separate events, it isn't a very rational position to take.
The same goes for dangerous places. One big dangerous place is far better than a thousand small ones, if only because it is easier to avoid.
See the area around Chernobyl for an example.
Yes, by all means, do.
Then you see the problem: a "normal fire, not unlike accidents we commonly see in the fossil fuel industry" was enough to render a huge area of land permanently uninhabitable and ruin thousands of peoples' lives.
Except that the area wasn't made "permanently uninhabitable" (but even if it was, that isn't such a big deal when you consider that 80% of the Earth's surface is perminantly uninhabitable to the same or greater degree). At worst, it's not quite as safe as it was, and won't be for another 30 years or so.
And as for ruining people's lives, it's a drop in the bucket compared to the use of fossil fuels (or the use of tobacco, or wars over oil, or the manufacture of pesticides, or...). To cite one such number out of context is simply FUD. Nuclear, per kilowatt hour, is still safer than the alternatives, is more scalable, and has less environmental impact.
That is why so many people are uncomfortable with nuclear power, and seek a better alternative.
I agree that the beliefs you outlined are responsible for the widespread rejection of nuclear power. But that doesn't make the beliefs sound--reality isn't a democracy, and many people believing something doesn't make it true.
(btw, I think the public's mind blames hydrogen more than helium ;^))
Oops. Typo. Thanks.
I think that "nuclear or coal" is a false choice.
Agreed. I'm a fan of space-based solar myself, but that's a topic for another day. The main advantage of nuclear is that it is available now, and is the only technology we have that is far enough along to prevent world war three from erupting over control of the oil. Conservation, wind power, etc. just don't have the scope to prevent that deadly showdown.
--MarkusQ
This is exactly the sort of thing I mean.
- It doesn't matter that there were no major health issues caused by Three Mile Island. It is still the primary reason for the souring of opinion toward nuclear power in the US because it was a mistake that nearly ended in disaster.
- It still exhibits the fundamental problem of nuclear power: one mistake and you're done. People do not believe it is wise to try to operate a facility with those kinds of tolerances.
- Something can be very safe, and still be considered too dangerous to use.
- For example, we don't use hydrogen in lighter-than-air vehicles anymore. It can be done safely for significantly less money than helium, and there were very few accidents. But you make a little mistake and "Oh the humanity!".
- It doesn't really matter that you only have a 1 in 6 chance of death in Russian Roulette. Eventually, you get a bad pull.
- And the truth is that no other form of energy can have such a long-lasting or damaging disaster.
- Hydroelectric dams come close, but you don't have to deal with millions of acres of radioactive land, just death and mud.
-- MarkusQSo the facts don't matter, the scare story is what counts. You do realize, of course, that almost anything that happens could be characterized as having "nearly ended in disaster" if you're willing to play fast and loose with the facts. So all you're really saying here is that the scare stories worked, which was my point.
Do these people by any chance drive cars? Because, you know, one mistake while you're driving and you're done. Same thing with using a fork.
Do you mean that people can (incorrectly) decide--or be persuaded--that something is too dangerous even though it is in fact safe? Or are you mired in some sort of Orwellian double speak, where "Safety is danger"?
Excellent analogy to Chernobyl, which was after all a normal fire, not a reactor problem, and not unlike some of the accidents that we commonly see in the fossil fuel industry. The Hindenburg disaster didn't have anything to do with the hydrogen, though it was widely blamed; the fault lay with the coating applied to the fabric and the same thing would have happened if it had been filled with helium. But even though this fact has been known for years, in the public mind it will always be helium's fault.
Not at all true. If you are engaged in other, risker activities your chances of death from the Russian Roulette go down. If, as in the case of nuclear, the risks are far lower than (for example) driving a car, your chance of death isn't effected much at all.
Tell me that when your front yard is hot enough to cook a pizza--or right, we'll all be dead then, won't we. I think the scare stories of both global warming and nuclear disasters are overblown, but to say that nuclear is worse is absolutely nuts. Read about surface conditions on Venus (which is very much like Earth, except it had run-away global warming and we haven't yet).
The other thing to consider: A nuclear disaster is something that would represent the most extreme failure mode of the system, while global warming is a natural consequence of the normal operation of the fossil fuel economy.
We have millions of acres of radioactive land right now. Pretty much anywhere there's granite is radioactive, for example. It may come as a surprise to you, but we get the radioactive material we use for nuclear power by digging it out of the ground. It's been there all along, and if we don't dig it up it will simply stay there, making the ground radioactive, just as its done for billions of years.
What I find ironic is how often people who don't trust the fossil fuel industry, and claim not to believe anything they say, etc. have been taken in by the anti-nuclear FUD spread by the very people they claim to distrust.
It's like some bad comedy routine.
Joe Public: I don't trust you.
Coal and Oil guy: I can understand that.
Joe Public: Nothing you can say will make me trust you.
Coal and Oil guy: I know just how you feel.
Joe Public: You do?
Coal and Oil guy: Sure. See that guy standing over there? The one with the pocket protector?
Joe Public: What, Nuclear Guy? Sure, I see him.
Coal and Oil guy: I don't trust him at all.
Joe Public: Why not?
Coal and Oil guy: He wants to kill all our babies and make giant insects and stuff.
Joe Public: Really?
Coal and Oil guy: Really. And he wants to make stuff that will kill people a bazillion years from now if they so much as think about it. That's why I don't trust him.
Joe Public: Wow. Thanks for the warning. But this isn't going to make me trust you any more than I did before.
Coal and Oil guy: I can understand that. Just so long as you don't trust him either.
Joe Public: Or don't worry about that. That guy is scary!
--MarkusQ
Two more points of agreement:
Back to the discussion:
There are a few problem with the "bouncing back and forth in time" explanation:
- We should see equal quantities of matter and anti-matter
- The amount of "memory" the magic particle would need would be astronomical (not a killer, since there's only one of them, but it seems inelegant)
- It conflates thermodynamic time, casual time, and time-as-a-dimension in a way that seems less than helpful here
The last point is the real kicker. If you allow the backward flow of information along world lines it seems to me that you loose the ability to compute intervals, and your whole space-time metric falls apart.(I'm aware that there are some "interactionist" models that accept this, but I don't recall much about them other than the assumption that the information flow across a unit of area is strictly limited to preserve the gross properties of the macroscopic world (e.g. we can't drive to the last century for the weekend) which would again give you the ability to measure something without affecting it. And thus even in them, there is unidirectional causuality of a sort, even if it isn't tied to a well-defined space time.)
If you don't allow backward flow of information, then being "strictly after" an event (delta_t > 0 and delta_t^2 > delta_x^2) should make it possible to measure it without affecting it.
--MarkusQ
A few things we seem to agree on
Now back to our regularly scheduled debate
You are misquoting me right after quoting me verbatim. My point was (and is) not that you would in effect be claiming that infinitely accurate measurement was impossible, but that all measurement was impossible. The position is logically untenable precisely because it reduces to an absurdity.
Say what? Please explain how, even in the absence of strict causality, when I measure (via spectral analysis) the temperature of a star that no longer exists (because, unbeknownst to me, it went nova millions of years ago) the temperature of my instruments can affect the temperature of the star, That doesn't sound like QM to me, it sounds like voodoo. No, I am quiting you verbatim from here.--MarkusQ
The 2-slit experiments I'm referring to neither require nor use a mirror. You're referring to something else.
The two-slit experiments you are referring to do not put (or claim to put) macroscopic objects in a superposition of states (which is what Schroedinger's cat is all about). So they don't prove (as you claimed) that this actually happened. The only experiments to date that I am aware of to actually test the issue did use mirrors (actually, a metal grid that served as a mirror at the wavelengths in question), for the reasons I outlined.
As for the temperature, there is no way that a thermometer of any size can't affect the sample its measuring unless it already is at the same temperature as the sample.
This really has nothing to do with QM (except in the popularized explanations), but in any case it both misses the point and is wrong.
For a clear example of how it is wrong, consider a thermometer that works by passively measuring the emission spectrum of the sample; in that case, it needs to be at the temperature of the background it is obscuring not at the temperature of the sample. But if it is far enough away even that won't matter, because the speed of light will prevent any influence from the thermometer on the sample.
To see why it misses the point, consider that, if you really meant it, you would in effect be claiming that all measurement was impossible. The real question is, does the thermometer affect the temperature enough to in turn affect the measurement it is making? This is a real question, of practical significance in many situations, and is the only part of all this that touches on QM. In the classical realm, you can always contrive a finer, more devious way to make the measurement to the desired accuracy while keeping the disturbance below the threshold. The key point of QM is that you can't do this indefinitely, that there is a quantifiable, universal limit to what is knowable without regard to any such "disturbance" by the measurement and thus not thwartable by clever design of instruments.
Every measurement, even in the macro universe, affects the thing measured. That this would NOT be the case at the quantum level, while counterintuitive at first, would upon reflection be surprising.
Not strictly true (for another example, I can attempt by various means to measure the distance Alexander the Great traveled during his life, but I cannot in any way affect it), but in any case this isn't QM, it's just an analogy used (misleadingly, IMHO) to explain QM (specifically the Copenhagen interpretation) without getting into the math.
Also, there is no need to resort to a superposition of states (which has always been a bit of a brainfuck imho) when there are better models. Superposition is only required if the time scale is unidirectional and can't be "rotated out" of the question and replaced by another vector, which has never been shown to be the case. The copenhagen gang lacked sufficient imagination to see the obvious.
There may be no need to resort to superposition in general, but the specific point in question here, "Do macroscopic objects, such as cats, enter into a superposition of states as claimed?" can not be dodged in this way. You said that of Schroedinger's cat (which asserts that macroscopic objects enter into a superposition of states) "Sounds weird, but its already been proven to be the case"; to now claim that this is "a bit of a brainfuck" while still maintaining that it is a proven, real effect doesn't wash.
-- MarkusQ
There are four boxes to defend liberty with: the soap box, the jury box, the voter box and the ammo box. Use in that order.
I think "they" got hold of the same list, and their response plan went something like this:
-- MarkusQ
It turns out, the cat is always either alive or dead, the mirror either broken or unbroken. Macro scale objects do not enter into a superposition of states in the real world. The assumed reason for this is that the larger the object, the more fragile (speaking very loosely here) a superposition of states is. By the time you get to a macro scale object at normal temperatures, you will never get a superposition--just moving an electron somewhere in another galaxy would be enough to collapse it.
On the other hand, the idea of macroscale objects in superposition is not nearly so fragile. It seems to be entrenched (paradoxically) because it's so counter intuitive.
--MarkusQ
P.S. As for the rest of your explanation, it's right in spirit but wrong in fact. Operating at the same level of analysis, you could just keep making smaller and smaller thermometers, or playing clever tricks with them to measure the temperature of the water indirectly, or without affecting it as much, so the argument as you present it begs the question.
The whole point of QM is that there is a bottom to all this, a point where you can't build a smaller thermometer, and no combination of clever tricks will help you. Your argument only seems to explain this because you silently assume it from the very beginning.
They aren't worried that RF noise will set off the explosives; they are worried that RF noise will interfere with the equipment they use to set off the explosives. They don't use a long roll of wire and a plunger style generator anymore--they set 'em off by radio. And they'd rather you didn't set them off by radio before they were ready.
-- MarkusQ
What in the heck is that supposed to mean?
--MarkusQ
Competition is good for consumers (and the market) under the assumption that they can make rational, informed decisions about the competing products. The whole purpose of the cell phone style pricing schemes (as with advertising, complex tax codes, etc.) is to make rational comparison so difficult as to be practically impossible, and thus invalidate the assumption; they are an attempt to subvert the process.
--MarkusQ
I suppose you'd like it even better if...
If I'm right, you may want to go cell phone shopping.
For my self, I far prefer a company that simply tells me what their product costs and then lets me decide if I want it.
--MarkusQ
As opposed to "we Linux users who are already running on 64 bit hardware and only bothered reading this thread to see what the fuss was about".
--MarkusQ
I strongly disagree.
The idea of a test being repeatable does not require "moving forward in time, and seeing the results"; it only requires that independent observers can replicate the results. Take astronomy, for example. If I have a theory that says "if you look carefully at a region of space that meets this specific set of criteria you should find more of this type of object than of this, by a ratio given in this formula" I can test it by going out and performing the observations. You can confirm it by doing the observations yourself, on the same volume of space or on another of you choosing. Neither of us has to have a Universe-Rewind button; we don't have to reboot the galaxy, or wait for a new one to form.
The same logic applies to geology, cosmology, etc. as well as biology, of which evolution is a keystone.
--MarkusQ
P.S. In any case, the test you propose ("A proper test of evolution would be to have some species, propose they will evolve a certain way, and then verify that they do indeed evolve that way.") has been done many times and consistently turns out to confirm the theory. So even if I agreed with your artificially constrained definition of "test" the original point still holds.
If they start to step by raising a leg on the far side, it will make them easier to tip.
If they raise a leg on the near side and try to step away, it will make them easier to tip.
Their best (and typical) response is to raise a near leg and move it towards the tipper, broadening their base and lowering their center of gravity. That, and only that, would makle them harder to tip. But that does not mean I agree with the articles conclusions.
--MarkusQ
Evolution is testable, falsifiable, and even demonstrable.
- Evolution makes some rather stringent predictions about the sorts of plants and animals we should expect to find, and the ones that we shouldn't. For just one example, we should expect to see animals that are systematically willing to die for a chance to mate, but not for something to eat or drink. And that is in fact what we do see. There are hundreds of such predictions, and they have all turned out to be correct,
- Darwin himself pointed out several ways in which his theory could be disproved, and many others have been discovered since. Yet for all the ways it could be falsified, it has not been.
- We can actually see evolution happening, in everything from domesticated animals to germs. Why do you think doctors worry about overuse of antibiotics?
--MarkusQ"First off, yes, it is a crime to leak the name of an undercover agent (or any other classified information) regardless of intent."
No, it is not.
Yes, it is. It's called "mishandling classified information" and you don't have to even leak it. Hiding notes about classified information in your socks suffices (remember Berger?). The whole point of having a system of classified information is that the people entrusted with the information aren't supposed to tell unauthorized people about it. Whether their intent is to impress their girlfriend, or burn a political adversary, or win a prize on a radio quiz show, or just make smalltalk while waiting in line at the theater doesn't matter.
As for the rest of your post, you might want to sniff the kool-aid you're drinking before you start calling names.
--MarkusQ
Let's do some "history correction".
Sure, lets.
Now this administration may be able screw up the invasion of the wrong country...
As opposed to Bill Clinton's invasion of two "wrong" countries Haiti and Kosovo... a "quagmire" I think we're still stuck in. Not to mention his poor execution of the efforts in Somalia and his indifference to the people of Rwanda.
I am no fan of Clinton, but you can't seriously be comparing the scope of Haiti and Kosovo to Iraq.
leak the names of CIA agents...
At this point unfounded speculation at best. Besides, it's not a crime to "leak" the names of CIA agents unless the intent was to expose them. In this case, it was hardly the intent to expose an undercover CIA operative (which Mr. Wilson's wife was not), but simply to disclose how Mr. Wilson got the assignment. But speaking of breaking laws, who was it that lied to a Grand Jury abou a blow job? Oh yes, that was Bill Clinton.
First off, yes, it is a crime to leak the name of an undercover agent (or any other classified information) regardless of intent. And yes, despite the administrations carefully worded talking points she was undercover, and the information was classified as "secret" in the memo the CIA provided to the WH.
And the "intent to disclose how Mr. Wilson got the assignment" fib has been sunk by the time lines--unless you are claiming they began an organized campaign to clarify a statement three weeks (mid June) before the statement was made (early July) and are intending to split hairs about the distinction between why Wilson in particular was send (selected by the CIA, after being suggested by his wife) from the real question of why anyone was sent on this particular assignment (do obtain more information, as requested by Cheney).
As for the "unfounded speculation" aspect, you may want to catch up on the news. We now know that the administration has repeatedly lied about this issue, including the claim that Rove & Libby had "nothing to do with it" which was changed to "first heard about her from reporters" and then to "were acting alone, not as part of any organized campaign" and that they were doing it "in response to Willson's NYT opinion piece" but started weeks before the piece was even written and did so in an amazingly unified and coherent fashion. We were told that "Cheney knew nothing about it," even though today we learn that Libby's hand written, dated notes of a meeting with Cheney in the days before the campaign started include the salient details.
But I guess all this overshadows the fact that the 9/11 commission says Mr. Wilson lied about the Nigeria-Iraq connection, which is what the liberals want.
Stripping the political baggage from your statement (facts don't care who "wants" them), Wilson was disputed on a single point; he said he "saw" that the documents were forgeries, but had not in fact personally "seen" the original documents. Understandable as a miscommunication, and hardly discrediting, especially as (IIRC) he clarified the point as soon as he was called on it. He has been proven correct and Dick "We know they have WMD" Cheney has been proven incorrect on every substantive point.
--MarkusQ
P.S. For the record, I was up in arms about Clinton and the BJ too. Both for the perjury and (perhaps more importantly) for the effect on his family. Hillary can take care of herself, but imagine the effect that must have had on his daughter. Not to mention that the Democrats would have been up in arms, crying sexual harassment if a CEO or the president of a university had done something similar.
But just as I hold the Democrats responsible for their actions, I expect the leaders of my own party to behave themselves in a way that brings credit, not shame, on the party. And this cabal of nincompoops is doing more damage to the Republicans than any Democrat could dream of doing.
Just as I turned in disgust from the latest Dan Lyons OSS flamebait (nominally about MySQL but mainly about his fear and loathing of Open Source) I thought to myself "What to I want to read now to get the taste out of my metaphorical mouth?"
I Dvorak flamebait? No, we just had one of those, and anyway reading two flamebaits in a row is bad for my blood pressure.
I know what would hit the spot! A Roland Piquepaille troll! We haven't had one of those for a while.
It's as if the gods heard my plea.
Darn those gods to heck anyway. I never thought I'd say it, but I miss Jon Katz!
--MarkusQ
I think we disagree about who is the victim. It isn't the person who's computer is taken over (I would call them an accomplice by virtue of negligence), it's the people against who the bot net was used.
Look at it this way; if there was a sudden fad for leaving loaded guns on the roof of your car when you parked, and street gangs were using them to commit crimes, would you just consider the people who left the guns on their cars victims? Their guns were stolen, after all. Or would you say that they shared some of the responsibility for the crimes?
--MarkusQ